You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It should be possible to optimise the following aggregation significantly by not requiring materialisation of the contiguous bitmap and by reusing intermediate containers during the aggregation (like FastAggregation.workShyAnd does):
// one API option:RoaringBitmapfoo(longmin, longmax, RoaringBitmap[] bitmaps) {
RoaringBitmapresult = newRoaringBitmap();
result.add(min, max);
for (RoaringBitmapbitmap : bitmaps) {
result.andNot(bitmap);
}
returnresult;
}
// another API option:RoaringBitmapbar(longmin, longmax, RoaringBitmap[] bitmaps) {
RoaringBitmapunion = newRoaringBitmap();
for (RoaringBitmapbitmap : bitmaps) {
union.or(bitmap);
}
RoaringBitmapresult = newRoaringBitmap();
result.add(min, max);
result.andNot(union);
returnresult;
}
It should be possible to optimise the following aggregation significantly by not requiring materialisation of the contiguous bitmap and by reusing intermediate containers during the aggregation (like
FastAggregation.workShyAnd
does):There would be a method added to
FastAggregation
:The text was updated successfully, but these errors were encountered: